FR2936673A1 - DEVICE AND METHOD FOR CONTROLLING THE QUALITY OF SERVICE OF THE BROADCAST OF AN ELECTRONIC SERVICE GUIDE - Google Patents

Abstract

The distribution of the ESG in good conditions and its conformity determine the quality of the user experience. Today, there is no specific tool for controlling and monitoring the delivery of a program guide service. The invention aims to solve this problem by proposing a device and a method for controlling and monitoring the broadcasting of an electronic program guide. More particularly, the invention defines a set of innovative criteria and their method of preparation. The determination of these criteria makes it possible to measure the quality of service of the broadcast of this ESG. In addition, these criteria are useful for diagnosing possible GSS diffusion issues. They also help to optimize the bandwidth used by the ESG broadcast.

Description

The present invention relates to the field of monitoring and evaluation of the quality of service of digital television broadcasting and more particularly the service of broadcasting of the electronic service guide or ESG (Electronic Service Guide in English).

The service electronic guide or ESG is defined, among other things as part of the broadcast of digital video DVB (Digital Video Broadcast in English) and more particularly DVB-H (Digital Video Broadcast Handheld) version of DVB for mobile terminals, in the technical specification documents of the European Telecommunications Standards Institute (ETSI) referenced TS 102 471 and an implementation guide is given in document referenced TS 102 592. It should be noted that the ESG is also use similarly in standards like OMA BCAST, DVB-SH, Media Flow, ATSC or MBMS, or other, the invention applies naturally to all these standards. This ESG consists of a data structure expressed in XML (Extensible Markup Language), an open standard of the W3C (World Wide Web Consortium). We are used to seeing this structure as a set of objects to describe all the services available within the distribution package and for each service all programs broadcast, broadcast schedules, information about program content and others. The ESG also contains broadcast or location information for the services that allow the terminal to locate and configure itself for the purpose of receiving the service. The ESG is composed of a set of XML fragments distributed via an IP (Internet Protocol) link, typically an IPDC (IP Data Cast in English) link, but any IP link available on the receiver can be used to broadcast all or part of the GSS. These fragments can be diffused according to a diffusion policy proper at the frequency of diffusion and update. They are a complex and large data structure. At the initialization of a receiving terminal, one of the first operations that the terminal must do is to load the ESG to be able to offer services to the user. Without GSS data, no service can be accessed. Therefore, we understand that the distribution of this ESG in good conditions and its compliance will determine the quality of the user experience. It will be crucial in all operations involving services such as the startup time required to access a first service, the time to switch from one service to another (zapping in English) and others. As the radio resource is limited, one of the major problems of service broadcasting is the optimization of bandwidth. Today, there is no specific tool for controlling and monitoring the delivery of a program guide service. The invention aims to solve this problem by proposing a device and a method for controlling the broadcast of an electronic program guide. More particularly, the invention defines a set of innovative criteria and their method of preparation. The determination of these criteria makes it possible to measure the quality of service of the broadcast of this ESG. In addition, these criteria are useful for diagnosing possible GSS diffusion issues. They also help to optimize the bandwidth used by the ESG broadcast. The invention relates to a method for controlling the broadcasting of an electronic service guide by a device for receiving a data stream comprising at least one electronic service guide called ESG composed of referencing fragments describing all the services. accessible to the terminal, including a step of connecting to a well-known broadcast address where an ESG start session is broadcast; an analysis step of this boot session to extract the location of the different ESG sessions; a step of connection and acquisition of different ESG sessions and localized disseminating fragments; a step of constituting the ESG from the fragments thus acquired as well as a step of calculating at least one quality of service criterion of the ESG broadcast. According to a particular embodiment of the invention, the method comprises a step of calculating a single criterion of the start-up ESG session, this criterion having a Boolean value according to the number of sessions broadcast on the Well-known address is one or another value. According to a particular embodiment of the invention, the method comprises a step of calculating an indicator of the number of erroneous corrigible fragments received.

According to a particular embodiment of the invention, the method comprises a step of calculating an indicator of the number of cycles required on average to correctly extract a fragment, the fragments being broadcast periodically within the sessions, the time of cycle being defined as the repetition time of the fragment.

According to a particular embodiment of the invention, the method comprises calculating such an indicator by type of objects. According to a particular embodiment of the invention, the method comprises a step of calculating an indicator corresponding to the ratio between the diffusion period of the FDT-type objects and the diffusion period of the other objects. According to a particular embodiment of the invention, the method comprises a step of calculating an indicator that reflects the consistency between a first descriptor that identifies the different ESG providers present on the IP platform and a second descriptor that indicates how to access the data of each of the providers described in the first descriptor within the boot session. According to a particular embodiment of the invention, the method comprises a step of calculating an indicator that has a value representative of the repetition time of a given object. According to a particular embodiment of the invention, the method comprises a step of calculating an indicator relating to the theoretical maximum acquisition time of the ESG. According to a particular embodiment of the invention, signaling information being included in the received data stream, the method comprises a step of calculating an indicator representative of the coherence between the IP addresses relating to the transmission of the data. ESG as advertised in the data flow signaling information and the actual addresses used for this broadcast. According to a particular embodiment of the invention, the method comprises a step of calculating a fragment richness indicator which measures for each fragment the number of fields filled out among all the possible fields. According to a particular embodiment of the invention, the method comprises a step of calculating a fragment completeness indicator which measures for each fragment the number of fields filled out among all the relevant fields, these relevant fields being a subset given possible fields. According to a particular embodiment of the invention, the method comprises a step of calculating an indicator of completeness of the ESG corresponding to the number of fragments having a reference on a non-diffused fragment.

According to a particular embodiment of the invention, the method comprises a step of calculating an indicator of completeness of the ESG corresponding to the number of isolated fragments not referenced by other fragments of the ESG and therefore useless. According to a particular embodiment of the invention, the method comprises a step of calculating a semantic coherence indicator indicating whether certain semantic features announced in the ESG correspond to the reality of the transmitted ESG data. According to a particular embodiment of the invention, the method comprises a step of calculating a list of unusable fragments with their reference number. According to a particular embodiment of the invention, the method comprises a step of calculating an indicator representative of the evolution over time of the loading of the ESG which indicates in real time the number of references to fragments not yet receipts discovered in the fragments already received. The invention also relates to a device for controlling and monitoring the broadcasting of an electronic service guide comprising means for receiving a data stream comprising at least one electronic service guide called ESG composed of a service tree. referencing fragments describing all the services accessible to the terminal; connection means to a well-known broadcast address where a start-up session of the ESG is broadcast; analysis means of this boot session to extract the location of the different sessions of the ESG; means of connection and acquisition of the different ESG sessions and localized disseminating fragments; means for constituting the ESG from the fragments thus acquired as well as means for calculating at least one service quality criterion of the ESG broadcast. The characteristics of the invention mentioned above, as well as others, will appear more clearly on reading the following description of an exemplary embodiment, said description being given in relation to the attached drawings, among which: Fig. 1 illustrates the software architecture of the IP layer in DVB-H; Fig. Figure 2 illustrates the GSS data model in the IPDC case; Fig. 3 illustrates the ESG broadcast chain; Fig. 4 illustrates the method for extracting the ESG and calculating the criteria according to an embodiment of the invention.

Typically, the ESG as described in ETSI TS 102 471, is broadcast within a DVB broadcast stream via the IPDC (IP Data Cast) data broadcast mechanism. This broadcast mechanism uses a periodic monodirectional diffusion (carousel in English). The broadcast protocol used is the protocol FLUTE (File Delivery over Unidirectional Transport in English defined by RFC 3926). Fig. 1 illustrates the software architecture of IP transmissions in a DVB-H context. Block 1.1 represents the DVB-H broadcast layer, therefore the transport stream. This transport stream consists of a set of MPE sections (Multi Protocol Encapsulation in English defined in ETSI document EN 301 192) carrying the IP packets of layer 1.3. Eventually, DVB-H communications can be supplemented by a point-to-point transport layer 1.2. Above the IP 1.3 layer, there are the conventional transport layers UDP (User Datagram Protocol in English defined by RFC 768) and TCP (Transmission Control Protocol in English defined by RFC 793). It is recalled that UDP is a datagram routing protocol with no mechanism for signaling and forwarding lost packets as opposed to TCP. UDP can therefore be implemented on a one-way link, it is lightweight, but does not guarantee the delivery of data. TCP requires an acknowledgment-receipt mechanism to guarantee delivery but making a bi-directional link mandatory. As a result, there is logically found the broadcasting of real-time content, for example of the audio and video type 1.9 implemented by real-time broadcasting protocols such as RTP (Real-time Transport Protocol in English defined by RFC 3550) and its protocol of RTCP control, 1.8 and 1.7, above UDP.

Just as logically, we find mechanisms based on the paradigm of queries and responses, typically bidirectional, such as repair mechanisms 1.14 to re-request fragments of services not or badly received via the HTTP protocol (Hyper Text Transport Protocol in English defined by RFC 2616) 1.13 above TCP.

The Service Purchase and Protection (SPP) protection and purchase services are partly implemented in the 1.6 unidirectional broadcast and partly via interactive mechanisms over TCP 1.15.

For the diffusion of data files having no real time constraint, one uses a diffusion protocol called FLUTE (File Delivery over Unidirectional Transport in English defined by RFC 3926) 1.10. This protocol allows the distribution of a collection of objects within sessions. These objects are advertised in one or more tables called File Delivery Table (FDT) which are themselves distributed as an object of the session. Each object can be diffused periodically and thus rebroadcast regularly. This protocol allows the broadcasting on a one-way link of 1.11 binary files and in particular the electronic program guide ESG 1.12. Although this is a one-way broadcast, the fact that the data is broadcast periodically makes it possible to repair an object badly received by a new reception delayed in time. Above, application layer 1.16 allows the implementation of broadcast services.

The electronic service guide consists of a tree of objects, referred to as fragments, referencing which describe a set of services, or bouquet, accessible to a terminal. It contains information to locate and configure the reception of services as well as semantic information about the content of these services. This object tree follows the data model shown in FIG. 2. This data model is described by an Extended Markup Language (XML) schema definition. The GSS is subdivided into ESG fragments, which can be instantiated in part from ESG. The indicated cardinalities correspond to the cardinalities specified in the XML schema definition. The bouquet fragment 2.1 (service bundle in English) specifies a set of services, understood as a grouping of objects offered to the user in the form of services. This bouquet can be used to link purchase information to bouquets. The purchase fragment 2.2 (purchase in English) contains the information to be displayed to the user to enable the purchase of the service.

The purchasing channel fragment 2.3 (purchase channe / in English) contains the interface allowing access to the purchasing system. The service fragment 2.4 describes the information relating to a service. The programming fragment 2.5 (schedule) specifies the time of broadcast of a service element.

The content fragment 2.6 (content in English) contains the meta data that describes the content regardless of the mode of dissemination of this content. The acquisition fragment 2.7 specifies the information to access a service or content.

This model is the model used in IPDC, variants of this model can be implemented in other neighboring protocols like OMA BCAST. According to this protocol, the programming and content fragments can be combined into one, for example. ESG sessions defined as the sessions used to broadcast ESG data are implemented as FLUTE sessions. When we talk about ESG, we are talking about a tree structure of ESG fragments. When we are interested in the transport sessions of these fragments we use more easily the term of object which is the term consecrated in the terminology of the protocol FLUTE. The different fragments are referenced using identifiers and version numbers. They are gathered in containers (containers in English) which themselves have identifiers and version numbers. It is these containers that form the objects broadcast in the sessions. In this document, the term fragment and object is used according to the context, the objects transported in the FLUTE sessions being here containers that can carry several fragments of ESG. By extension, we often talk about the fragments transported in the FLUTE session without always explaining the encapsulation in the containers. The diffusion of this guide of the services is thus done by the diffusion in the form of a session FLUTE of all the fragments constituting this guide. These fragments are diffused periodically in a period that can vary depending on the fragments. The IP packets constituting this broadcast are encapsulated in MPE sections inserted into the broadcast transport stream. They usually come in either burst into burst (English) services or in the form of specific ESG bursts. Fig. Figure 3 illustrates the typical DVB-H service broadcast chain. A server is responsible for producing the ESG, it is the server 3.1 of the figure. A set of servers will produce the different services and encode them, it is the servers 3.2, 3.3 and 3.4 in the figure. The ESG and the encoded elementary streams of the services are then sent to a 3.5 equipment called IPE (IP Encapsulator in English) responsible for encapsulating and multiplexing these elementary streams. It is this equipment that generates the MPE sections and bursts from the received IP packets for the different services. The IPE 3.5 thus generates as output a single multiplexed stream containing the various services, the unrepresented signaling information and the service guide. This stream is routed to at least one broadcast point 3.6 to be broadcast to terminals 3.7, 3.8 and 3.9. The method for controlling and monitoring the broadcast of an electronic service guide is therefore typically implemented by one of the terminals 3.7 to 3.9. It may be a generic user terminal with a specific program enabling it to implement the method or a terminal dedicated to the flow test. This terminal is therefore a device for receiving a data stream comprising at least one electronic service guide. This guide is transmitted via an IP transport channel, typically the IPDC file broadcast channel, but it can be any IP transport channel such as a WiFi channel or others. The acquisition of an ESG by the receiving terminal comprises several steps illustrated in FIG. 4. In a first step 4.1 the terminal connects to a well-known IP address (well known address). At this address is broadcast an ESG start session which contains the announcement of the different ESGs present on the IP platform. In a second step 4.2, this startup session is analyzed to extract the location, that is to say the broadcast address, of the different ESG sessions. In a third step 4.3, the terminal connects to each of the announced addresses to acquire the ESG data that are broadcast there. We then acquire the different sessions ESG and localized. In a fourth step 4.4, the data thus acquired are used to inform a tree structure of objects constituting the electronic guide services. One or more ESG broadcast QoS criteria are calculated in a step 4.5. These calculations use data that can be retrieved during one or more of ESG acquisition steps 4.1 to 4.4. ESG is required for several applications on the terminal. First, when the terminal is initialized, it is necessary to acquire the ESG to be able to offer a list of services accessible to the user. He can then choose a service. The terminal then finds in the ESG the information it needs to locate and receive this service. It can be seen that the reception time of the ESG is determinant in the time between the switching on of the terminal and the moment when the user is able to start the consultation of a service. ESG information is also used during a service change operation (zapping). In addition, the semantic information of the ESG will allow the user to clarify his choice. This information forms attractors bringing the user to the consumer. For all these reasons, a good and reliable reception of the ESG is an element of the quality of service felt by the user of the terminal. As much as the criteria allow today to determine the overall quality of DVB broadcast, we can cite the technical report of ETSI TR 290, as no criterion or device can measure a specific quality of service of the dissemination of ESG. We will therefore define a method for estimating the quality of service of the electronic service guide. This process is based on the definition of a set of quality of service criteria of the ESG broadcast and the associated measurement method to obtain reliable information on the dissemination of ESG. A quality of service criterion of the GSS diffusion is therefore defined as a criterion calculated during the acquisition of the GSS and relating to the quality of service of this broadcast. It is thus a criterion specifically related to the GSS, its conformity or its diffusion. These criteria can be used not only to measure this quality of service, but also, at least for some of them, to help in the diagnosis of this diffusion. Indeed, some of these criteria can help to understand the source of possible malfunctions or slowness in this diffusion.

Advantageously, these criteria can be classified according to a hierarchy of severity levels, allowing to filter and select only certain levels of severity. For example, it is possible to define a first level of severity called critical for malfunctions that can prevent the extraction of ESG. A second level will be called a warning to indicate malfunctions that will slow the extraction without preventing it. A third level said information will relate to malfunctions reducing the comfort of the user, but not hindering the extraction of the guide, for example, by limiting the information available on a service. They are calculated cyclically and some give a statistical estimate of a measurement and for others a validity status of a quality criterion. These can be used directly as an alarm. For others, minimum and maximum thresholds can be set to define dysfunction limits. The calculation of these criteria can take place in each of the steps described in FIG. 4 according to the calculated indicators.

A criterion of uniqueness of the startup FLUTE session (bootstrap in English) can thus be defined. In fact, according to the ETSI TS 102 471 standard, the acquisition of the ESG starts with the acquisition of data broadcast in a single start-up FLUTE session available at a well-known address.

This well-known address is the 224.0.23.14 multicast IP address in IP version 4 and ffOx: 0: 0: 0: 0: 0: 0: 12d in IP version 6 on port 3937 or on the port 9214. Also, a receiver may admit that the first FLUTE session it detects at that broadcast address is the ESG start session. As a result, the presence of multiple FLUTE sessions at the well-known IP address, such as the absence of a boot session, can disrupt the acquisition of the ESG or even make it impossible. Advantageously, this criterion is therefore assigned a critical severity level. The method of determining this criterion is to listen to the data broadcast on the well-known address to determine the number of FLUTE sessions broadcast at this address. This criterion has a Boolean value depending on whether the number of sessions served on the well-known address is one or another value. This number is determined by analyzing ALC / LCT packet headers used in FLUTE sessions. These headers contain an indicator called the Transport Session Identifier (TSI) that uniquely identifies the FLUTE session among sessions from a given source. Each session is thus uniquely determined by the source IP address pair present in the IP and TSI header present in the ALC / LCT header. It can therefore be verified that the FLUTE packets received on the well-known broadcast address correspond to a single source address pair and TSI to calculate this criterion.

Occasionally, errors occur when receiving objects sent via a FLUTE session. The latter protocol being unidirectional, no information on the state of the receiver is sent back to the server. So there is no error notification. However, erroneous objects can be repaired if an FEC code is used, and is able to correct errors. Otherwise, the receiver must wait for the next cycle for another attempt to extract the desired data. We must distinguish two cases, in the first case the fragment received is repairable, but the repair by one of the mechanisms above causes a delay slowing the acquisition of ESG. In this case, we are dealing with a malfunction of the warning level. In a second case, the repair is not possible, the fragment is not repairable and is lost. In this case, we are in the presence of a dysfunction of the critical level. The invention proposes two types of indicators for monitoring this phenomenon of data loss, a first to give the rate of erroneous objects corrected by the FEC code, and a second which gives the number of cycles required on average to retrieve an object. . It is thus possible to deduce the number of cycles necessary when the FEC code is not used. The invention defines a first indicator representative of the number of erroneous objects that can be corrected. It will depend on criteria, which are the number of objects received, the presence of errors on reception, the presence of an error correction code (FEC) and finally the possibility of correcting or not the errors found. This indicator can be calculated according to the following formula: number of erroneous objects correctable * 100 number of received objects where the number of objects received corresponds to the total number of ESG containers received and the number of corrected erroneous objects corresponds the number of fragments received with at least one error and that could be corrected by one of the three repair methods, correction code, repetition or repair mechanism. The invention defines an indicator representative of the number of cycles required on average to correctly extract an object. It will depend on criteria, which are the number of objects received, the presence of errors on reception, the presence of an error correction code (FEC) and finally the possibility of correcting or not the errors found. To calculate this indicator, a particular acquisition mode is used. In this mode, one seeks to extract all the objects received even when the object, because of the periodic diffusion, is received again and even if it has already been correctly extracted during a previous passage. In this mode, if we define the number of objects received as the sum of the reception occurrence numbers of each object, each of these occurrences giving rise to an extraction attempt, we can then express in number of cycles by object the indicator and calculate it according to the following formula: number of received objects c = number of extracted objects It is expressed in cycles by objects, the measurements to be made on a time corresponding to several periods of diffusion of the objects to be meaningful.

The announcement of the objects present in the FLUTE session via the FDT may not be complete, it is advantageous to offer these indicators by type of objects. For example, these indicators can be calculated for FDT type objects, for objects mentioned in the FDT or for objects present in the session, although not described in the FDT. These metrics provide valuable information about session compliance, and can help diagnose problems in session generation. Advantageously, an indicator corresponding to the ratio between the diffusion period of the FDT type objects and the diffusion period of the other objects generally identical for the other objects will also be measured. Indeed, the period of diffusion of the FDT influences the time necessary to its good reception and thus to the good reception of the other objects of the session. For these indicators, the user of the control device is fixed or allowed to set a window of validity of the indicator. The user therefore sets lower and upper bounds for the value of the indicator. When the indicator goes out of the validity window, an alarm is generated. This alarm may appear visually on the device screen. During the startup phase of 1ESG (bootstrap operation), two descriptors are delivered. A first descriptor, ESGProviderDiscovery that identifies the different ESG providers present on the IP platform. A second descriptor, ESGAccessDescriptor, which indicates how to access the data of each of the providers described in ESGProviderDiscovery. If a provider described in the first descriptor does not have an ESG entry - ESGEntry - in the second, its data can not be accessed. Similarly, if the number of ESG entries present in the second descriptor is less than that of the first descriptor, the data can not be extracted from all the ESG providers present on the IP platform. Thus, a binary criterion is defined that reflects the consistency between these descriptors of the ESG start session, the start session being the session broadcast on the well-known start address of the ESG.

This criterion is analyzed during the reception of this start session by comparison between the two descriptors. This criterion has a critical severity level because an inconsistency may mean that at least one full pan of the GSS is not accessible.

The objects of the ESG are transmitted via the FLUTE sessions in a cyclic so-called carousel mode. The extraction times of all the data linked to an ESG may vary according to the choice of bit rates, that is to say the period between two diffusions of the objects of the different FLUTE sessions, hence the need to monitor this period for the different objects, and especially for the most critical objects. Thus, we will monitor more particularly: - The ESGProviderDiscovery descriptor, the standard TS 102 592 recommends a repetition time around 10s. - The descriptor ESGAccessDescriptor, the standard TS 102 592 recommends a repetition time lower than 1s. - The FLUTE session of the ESG announcement carousel, which must be broadcast at least once per DVB-H burst (DVB-H burst in English) according to the standard TS 102 592. Each of these indicators therefore has a value representative of the repetition time or diffusion period of the object concerned. Permissible value limits, a minimum and a maximum, can be set to generate alerts. With regard to the FLUTE session of the ESG carousel, the indicator can be defined as the ratio between the number of broadcasts of this announcement session and the number of DVB-H bursts received during a given time, this ratio to be greater than or equal to 1. Advantageously, it is also possible to create an indicator relating to the theoretical maximum acquisition time of the ESG. Indeed, we know that to acquire the ESG we must first acquire the FDT objects that reference the fragments of ESG broadcast. When one knows the period of diffusion of the FDT and these other fragments, one can deduce a minimum time of acquisition. This theoretical time can be the subject of an indicator. This indicator can be used, on the one hand, as an indicator of the quality of service of this ESG. The lower it is, the faster the acquisition is theoretically fast and therefore the quality of service good. On the other hand, one can also compare the actual ESG acquisition time to this theoretical time. The actual acquisition time may increase compared to the theoretical time because of transmission problems resulting in a high need to use the repair service which slows the acquisition. The theoretical time as the real time of acquisition can also be influenced by the ability of the terminal to acquire FLUTE sessions in parallel.

It can be seen that one can, on the one hand, define a criterion relating to the total acquisition time of the ESG and the criteria relating to the acquisition of certain specific fragments. These specific fragments are not limited to the given examples. Advantageously, the user interface makes it possible to select the fragment or fragments for which it is desired to obtain an indication relating to the acquisition time of said fragment. Among the signaling tables defined in ETSI TS 102 470 entitled Digital Video Broadcasting (DVB); IP Datacast over DVB-H: Program Specific Information (PSI) / Service Information (SI) it is one defined in section 4.8 and called INT (IP / MAC Notification Table in English) which announces the IP platforms and for each of them they all the IP addresses used. We must therefore find all addresses used for ALC / LCT channels used by FLUTE sessions. An indicator representative of the notification of the target IP addresses of the ALC / LCT channels in the table INT is thus defined. This flag is set when the address of an ALC / LCT channel used for ESG broadcast is not in the INT table. Advantageously, it is also verified that all the addresses present in the INT table and announced as relating to the broadcast of the ESG do indeed correspond to effective broadcasts. An indicator representative of the coherence between the IP addresses relating to the broadcast of the ESG as announced in the signaling information of the data stream and the addresses actually used for this broadcast is thus calculated. All criteria are applicable to both the IPDC mechanism and the OMA BCAST mechanism. A criterion of richness of the fragments is also defined. We have seen that the ESG consists of a tree of fragments referencing each other. A fragment is therefore a data structure whose fields are references to other fragments according to the data model, except for fragments constituting leaves of the tree. The comfort of the user depends on the degree of filling of these fields of fragments. It can also go for the completeness of the GSS.

A criterion can therefore measure for each fragment the number of fields filled out among all the possible fields. An average value can be established for the entire GSS. This indicator is representative of the rich content of the ESG and therefore the level of information it provides to the user. To calculate this indicator, we define a first value representative of the set of possible fields for a fragment i called Cpi and a second value representative of the number of filled fields of the fragment i called Cri. We can then define the criterion according to one of the following formulas:

(Cpi ù Cri) L Cri r = or r =; Cpi LCpi The first formula gives the rate of unfilled fields while the second gives the rate of filled fields.

Advantageously, since the number of possible fields defined by the standard can be high and some of these fields prove not really relevant in a given application, the user is given the possibility of defining for each fragment a set of so-called fields. relevant. In this case, this set of relevant fields replaces the set of possible fields in the calculation formulas of this indicator.

There is also another criterion relating to the completeness of the GSS as a whole. Indeed, it can be found that fragments refer to other fragments that are not scattered or missing fragments. These missing references can have several consequences of various gravities. On the one hand, they can lead to a lack of additional information relating to a service. In this case, we are faced with a criterion of information severity level. But they can also make the fragment with the missing reference unusable. One can cite the case of a content not linked to information making it possible to buy it. In this case the severity level is critical. In addition, the fragment, although broadcast, but rendered unusable by the lack of a referenced fragment may also be considered missing and cause a problem with the use of another referencing fragment. It is thus seen that the non-diffusion of a single fragment can cause, step by step, the loss of use of an entire branch of the tree constituting the ESG.

Advantageously, semantic information on the missing fragments can be taken into account here to determine the level of severity assigned to the criterion. As already seen above, the absence of a referenced fragment does not have the same effect depending on the nature of the fragment. In addition, it is also possible in this case to refine the attribution of a level of severity and to differentiate the impacted applications. The absence of the same fragment can be critical for the application of change of service (zapping in English) but does not prevent access to a first service at startup. We assign, therefore, the level of severity depending on the application such as startup, change of service, podcast (podcast), advertising channels, POI (Point of Interest) broadcast for applications of navigation or others.

On the other hand, we can also observe the scattering of fragments that do not refer to any fragment of the GSS and that are not referenced by any fragment of the GSS. These isolated fragments are therefore not usable and although they do not disturb the acquisition and functioning of the valid ESG fragments, these fragments are scattered unnecessarily and cause a loss of bandwidth. They could usefully be removed from the stream. This analysis of the isolated fragments must be extended to a set of fragments not related to the general tree of the disseminated GSS. This useless set is characterized by the fact that no fragment of the ESG refers to any of the fragments of the set, and that the only references present in the fragments of this set refer to other fragments of the set. exclusion of fragments of disseminated GSS. Two new criteria are thus defined corresponding to the number of fragments having a reference on a non-diffused fragment and to the number of isolated fragments not referenced by other fragments of the ESG and therefore useless. These indicators are calculated by analysis of the fragments received once it is estimated that the entire GSS must be received. It should be noted here that there is a difficulty in determining when it is estimated that the entire GSS is received. In practice, a duration chosen by the user for the calculation of these indicators is determined. Other semantic consistency criteria can be defined. These criteria indicate whether certain semantic features announced in the GSS are consistent with the reality of the ESG data transmitted. For example, we can define a criterion relating to conditional access to content that verifies that when a service is advertised as using an access control system, the content associated with this service is encrypted according to the encryption methods linked to it. to this access control system. When the terminal has different IP links, for example a Wifi link or a 3G link, the ESG broadcast in the stream may be supplemented by ESG data received via the complementary IP link. In this case, advantageously, these coherence criteria of the ESG make it possible to test the coherence between the ESG data received by the different links. Advantageously one can also extract the list of unusable fragments with their reference number and possibly their type.

Advantageously, a criterion representative of the evolution over time of the loading of the ESG is also determined. This criterion is defined as indicating in real time the number of references to fragments not yet received discovered in the fragments already received. This indicator is initialized to 0. When receiving a fragment, it is reduced by 1 if this fragment is one of the fragments whose reference was found and which had not already been received. It is increased by the number of references to fragments not yet received contained in the fragment. It is seen that this indicator must increase at the beginning of the ESG reception to stabilize at the value of the number of references to missing fragments over time. Its real-time display makes it possible to visualize the progress of the acquisition of the ESG and to determine a moment at which it is considered that the acquisition of the fragments broadcast is complete. We can also set a deadline and consider that if the value of this indicator does not change for a period greater than or equal to this period, we consider the acquisition as completed. The previous indicators for the completeness of the GSS are then calculated.

Although described mainly in the context of a broadcast of the ESG via the IPDC mechanism of DVB-H, the invention adapts to other modes of diffusion such as OMA BCAST, directly for most criteria or by means of some obvious adaptations for the skilled person. It can also apply when the ESG is broadcast over other IP channels, such as a Wi-Fi broadcast or other IP link. The ESG typically includes several ESGs each dedicated to a particular operator. Advantageously, the invention makes it possible to select one of these ESGs and to calculate the criteria only for this ESG.

Claims (1)

CLAIMS1 / A method for controlling the broadcasting of an electronic service guide by a device for receiving a data stream comprising at least one electronic service guide called ESG composed of referencing fragments describing all the services accessible to the terminal , comprising the following steps: - a step of connection to a well known broadcast address where is broadcast a boot session of the ESG; - an analysis step of this startup session to extract the location of the different sessions of the ESG; a step of connection and acquisition of the different sessions of ESG thus localized diffusing fragments; a step of constituting the ESG from the fragments thus acquired; Characterized in that it further comprises: - a step of calculating at least one quality of service criterion of the GSS broadcast. 2 / A method according to claim 1, characterized in that it comprises a step 20 of calculating a single criterion of the boot ESG session, this criterion having a Boolean value according to the number of sessions broadcast on the well known address is one or another value. 3 / A method according to claim 1, characterized in that it comprises a step 25 of calculating an indicator of the number of erroneous corrigible fragments received. 4 / A method according to claim 1, characterized in that it comprises a step of calculating an indicator of the number of cycles required on average to correctly extract a fragment, the fragments being broadcast periodically within sessions, the cycle time being defined as the repetition time of the fragment. 5 / A method according to claim 4, characterized in that it comprises the calculation of such an indicator by type of objects.6 / A method according to claim 1, characterized in that it comprises a step of calculating a indicator corresponding to the ratio between the diffusion period of FDT type objects and the diffusion period of other objects. 7 / A method according to claim 1, characterized in that it comprises a step of calculating an indicator that reflects the consistency between a first descriptor that identifies the different ESG providers present on the IP platform and a second descriptor which indicates how to access the data of each of the providers described in the first descriptor within the boot session. 8 / A method according to claim 1, characterized in that it comprises a step of calculating an indicator that has a value representative of the repetition time of a given object. 9 / A method according to claim 1, characterized in that it comprises a step of calculating an indicator relating to the theoretical maximum acquisition time of the ESG. 10 / A method according to claim 1, characterized in that signaling information being included in the received data stream, it comprises a step of calculating an indicator representative of the coherence between the IP addresses relating to the broadcast of the ESG as advertised in the data flow signaling information and the actual addresses used for this broadcast. 11 / A method according to claim 1, characterized in that it comprises a step 25 of calculating a fragment richness indicator which measures for each fragment the number of fields filled out among all the possible fields. 12 / A method according to claim 1, characterized in that it comprises a step of calculating a fragment completeness indicator which measures for each fragment the number of fields filled out among all the relevant fields, these relevant fields being a subset. given set of possible fields.13 / A method according to claim 1, characterized in that it comprises a step of calculating an indicator of completeness of the ESG corresponding to the number of fragments having a reference to a non-diffused fragment. 14 / A method according to claim 1, characterized in that it comprises a step of calculating an indicator of completeness of the ESG corresponding to the number of isolated fragments not referenced by other fragments of the ESG and therefore useless. 15 / A method according to claim 1, characterized in that it comprises a step of calculating a semantic coherence indicator indicating whether certain semantic features announced in the ESG correspond to the reality of the ESG data transmitted. 16 / A method according to claim 1, characterized in that it comprises a step of calculating a list of unusable fragments with their reference number. 17 / A method according to claim 1, characterized in that it comprises a step of calculating an indicator representative of the evolution in time of the loading of the ESG which indicates in real time the number of references to fragments not still received discovered in the fragments already received. 18 / Device for controlling and monitoring the broadcasting of an electronic service guide comprising: means for receiving a data stream comprising at least one electronic service guide called ESG composed of a tree of fragments. referencing describing all the services accessible to the terminal; means of connection to a well-known broadcast address where a start-up session of the ESG is broadcast; analysis means of this start-up session to extract the location of the different sessions of the ESG; means for connection and acquisition of the various ESG sessions thus localized, broadcasting fragments; means for constituting the ESG from the fragments thus acquired; characterized in that it further comprises: - means for calculating at least one quality of service criterion of the GSS broadcast.